CrossKD: Cross-Head Knowledge Distillation for Dense Object Detection

CrossKD是一种用于密集目标检测的知识蒸馏方法，它通过在目标检测的分类头上应用知识蒸馏（KD），从而实现对目标检测的位置头的定位蒸馏（LD）。该方法的目的是通过将分类头上的知识传输到定位头上，以提高密集目标检测的性能。<span class="em">1</span><span class="em">2</span><span class="em">3</span>
#### 引用[.reference_title]
- *1* [CrossKD 原理与代码解析](https://blog.csdn.net/ooooocj/article/details/131628652)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v93^chatsearchT3_1"}}] [.reference_item style="max-width: 50%"]
- *2* *3* [目标检测-定位蒸馏：logit蒸馏与feature蒸馏之争](https://blog.csdn.net/c9Yv2cf9I06K2A9E/article/details/124395112)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v93^chatsearchT3_1"}}] [.reference_item style="max-width: 50%"]
[ .reference_list ]